Chapter 1 Overview

The goal of this project is to maximize the profit with acceptable default risks. We built a model to study key variables that might lead a customer to default; then, we referred to the rejection data to adjust the model’s accuracy. Based on the final model we created, we designed the scorecard, which measures the applicants in three dimensions, Persons Per Household, Name of Credit Card, Time at Job (months). The results showed that to keep the risk of default rate at 3.23%, our model can help the Bank accept 96% of applications, and to keep the acceptance rate at the original 75%, the model can minimize the default rate to 1.41%. We suggest the company use our model to assess credit card applications and set the score cutoff between 441 and 472 to lower the default risks while accepting more applications.

Scorecard